|
|
|
|
|
| Management Briefings
|
|
|
|
|
|
When worlds collide: SOA and legacy: Martin Rice, Erudine (June 2010)
|
|
|
|
|
|
In this fast-changing business world, one of the main areas of concern is the strain on the
services delivered by existing legacy IT systems. Moving from a batch to a web services world,
particularly as the shift to cloud-based services accelerates, poses many challenges.
In particular, the traditional end-to-end systems that still dominate the IT infrastructure of many
established businesses are the antithesis to re-use, with individual functions lost and interfaces
obscured within millions of lines of code.
While creating new interfaces between the underlying legacy system and the end user through
a new architectural approach may address the short-term business need, an SOA-led approach
doesn’t tackle the underlying challenge that when systems are written, they become legacy.
Yet if this underlying challenge is not overcome, then the promised efficiencies and flexibility of
cloud computing will not materialise and the shift to re-usable services will be extremely painful.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Building an architecture: Mark Couzens, SELEX SI (March 2010)
|
|
|
|
|
|
Why should companies use an enterprise architecture (EA) strategically?
Primarily, it’s because large systems integration and engineering projects frequently involve
upgrading or replacing mission-critical business information systems. Yet these systems have
often emerged in an organic, tactical manner to solve local business problems, and may
represent significant organisational internal investment. Unfortunately such a development
lifecycle often creates a ‘disintegrated’ system or system of systems when viewed from the
strategic perspective.
In order to mitigate business risks and provide a foundation for delivering future strategic
change, businesses need to develop a detailed, concise and robust description of their own
organisation – and an effective and industry-proven method of doing this is via an enterprise
architecture.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Many hands...: John Abram, Monetical (February 2010)
|
|
|
|
|
|
An increasing number of software development heads are finding they are managing a highly distributed organisation – not as
a result of any operational objective but as a consequence of commercial mergers and acquisitions, increased partnership
activities, or enforced short-term cost reductions.
The result is often a dramatic reduction in enterprise development performance caused by two significant changes: decline in performance capability; and erosion of development expertise.
Combined, these two problems greatly impact an organisation’s ability to identify and introduce the right changes to its
development environment to reverse the performance decline. So how do you find the best performance turnaround solution
for your business?
The first step is to quantify your company’s current performance capability and the complexity of your environment. This will
prove highly valuable when developing a business case for investment and measuring the return on investment following
performance turnaround, as well as demonstrating to executives the underlying reasons for any decline in performance.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Twins separated at birth: Tony Baer, Ovum (September 2009)
|
|
|
|
|
|
One of the main purposes of service oriented architecture (SOA) is to slice through software silos in order to help organisations
grow more agile in automating their business processes. Effective governance is essential if SOA is to deliver on this promise.
However, in their zeal to manage SOA, many organisations are at risk of spawning yet another new governance silo that
becomes a ‘special case’ for the software development lifecycle (SDLC) portion of the application lifecycle.
The uniqueness of SOA is attributable to the fact that, while it requires skills for working with common programming languages
and platforms, it also demands specialised architectural skills or awareness and, in some cases, specialised tools or processes
that parallel or fork from the SDLC.
In the long run, relegating SOA to the software development organisation as a special case creates the very duplication that
SOA itself was supposed to eliminate.
Mainstreaming SOA as a realistic option for enterprise software developers requires that SOA be reconciled with the SDLC
and full application lifecycle, rather than run as a parallel special case.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Making SOA pay: Paul Heaviside, PA Consulting (June 2009)
|
|
|
|
|
|
Just a few years ago, amid much fanfare and vendor activity, service oriented architecture
(SOA) burst into the hearts and minds of IT departments across the globe. As suppliers
invested heavily to release new products – or at least ‘SOA-enable’ their existing offerings –
they convinced IT departments to buy and thus reap the rewards SOA promised.
Now some years on, it is safe to say these investments have met with varying degrees of
success.
Focusing on their ESBs and XSDs, many IT departments have failed to deliver the promised
returns from SOA. And the reputation of the technology has suffered. Yet in most of these
cases, it is not SOA that has failed, but the application of SOA.
Like a dotcom boom or a financial bubble, even some of the most experienced IT professionals
forgot some basic principles when it came to their ‘SOA investment’ – it’s all about the
business!
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Adopting agile: Dorothy Tudor, TCC (April 2009)
|
|
|
|
|
|
Agile development has evolved since the mid-1990s as a reaction against ‘heavyweight’ methods such as the waterfall model
of development.
An agile approach is one that delivers business-focused results quickly and effectively. It should do this reliably, predictably
and within a tightly constrained budget and a specific timeframe.
An agile method generally promotes incremental development and delivery. It will advocate simplicity, it should be easy to
learn and use, and sufficiently well-documented to be teachable and repeatable. It should also be adaptive, allowing for
changes in requirements occurring during the development cycle whilst still controlling ‘scope creep’.
It should offer a set of best practices that allow for rapid delivery of high-quality product; a leadership philosophy that
encourages team work, empowerment and accountability; a team-based approach, requiring close customer/developer
collaboration; and a business-focused approach that aligns development and delivery with customer needs and company
goals.
Finally, it should also promote a sustainable pace of working and visibility of progress.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Dodging the recession: W Emmerich & C Bröcker, Zuhlke Engineering (February 09)
|
|
|
|
|
|
An economic downturn has many effects on companies’ IT strategies in general, and their enterprise integration in particular.
There is an increased focus on cash preservation through reducing inefficiencies in organisations. Companies may also need
to reorganise quickly and close business divisions, services and product lines.
In addition, the rate of mergers and acquisitions is accelerating as a result of share price falls that enable stronger
organisations to acquire weaker ones. Organisations may also have to respond to the liquidation of companies, for example by
adjusting their supply chains.
In a recent article in the Financial Times, Donald Sull argued that a period of economic hardship is “an ideal opportunity to
implement change and instil better practice”.
In light of this, a modern enterprise integration approach can help organisations in addressing such change. It can in fact
provide an IT system architecture that reduces manual integration work, supports internal reorganisations as well as mergers
and acquisitions, and helps organisations to adjust their supply chain. So how can this be done?
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Living legacy: Shantanu Bhattacharya, Siemens (November/December 2008)
|
|
|
|
|
|
Many companies want to introduce a service oriented architecture (SOA) into the organisation
to make their processes flexible, adaptable and supple. But they already have a set of systems
in use for their business processes.
So how do you integrate SOA with your legacy applications to get more value out of them? This
article takes you through the steps you need to make it happen – and the pitfalls to avoid.
It’s clear that legacy systems form the base from which an enterprise can continue to improve
on its existing processes. So any strategy to bring in value without incorporating legacy systems
will have limited success.
Some legacy systems can be integrated with SOA relatively easily, others with more difficulty.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Transformation trilogy: Nigel Devenish, Altkon (August 2008)
|
|
|
|
|
|
SOA – service oriented architecture. BPM – business process management. EDM – enterprise decision management.
These three technologies have all been developed separately as business improvement tools. But bring them together and
they form the basis for sound governance, risk and compliance at all levels of an organisation, creating a powerful ROI
argument.
In fact, the return is even better than it might first appear. The ROI model is not so much the sum of the parts – eg, SOA +
BPM + EDM = ROI. It is more the multiplier SOA x BPM x EDM = ROI.
Whether the motivation is compliance, efficiency, productivity, profitability, management and control – or combinations of all
these – it is a powerful message. But when should companies integrate these three approaches and how? The promise offered by the combination of SOA, BPM and EDM is partly down to market maturity. These tools have all been
available to early adopters, but until now the costs for the individual technologies have been out of reach for all but the Tier-1
organisations.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Future perfect?: Peter Boggis (May 2008)
|
|
|
|
|
|
The continued global consolidation of industries and the accelerating competition from emerging economies are driving
significant change in the businesses we used to understand how to manage. As one global CIO recently said to me: “The
future is clearly no straight extrapolation of what’s worked in the past. We must stop trying to drive the car through the rearview
mirror. What’s made us successful in the past is no longer sufficient to make us successful in the future – radically
different capabilities are required of us as leaders for the future.”
One way of navigating towards a future state which may be very different from the past or present is through an enterprise
architecture or ‘target operating model’ for the company.
Within global, multi-company groups, the desire to demonstrate that ‘the whole is greater than the sum of the parts’ has led to
a quest for synergies across geographies, products/services, business processes, infrastructure and technology solutions. For
many organisations this is not a new journey – companies such as Cisco, Kraft General Foods or GE have all been active in
this area for a number of years.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
BPM: beyond technology: Neil Ward-Dutton, MWD (March 2008)
|
|
|
|
|
|
Business process management is a hot technology. But, as I explained in my first article, thinking about BPM purely in terms of
technology is a sure way to confusion and, ultimately, failure. That first article explained the role of BPM technology – this
follow-up focuses on how BPM can be applied to solve business problems.
Business processes exist at a number of levels within organisations: operational processes which handle particular units of work within a business activity – for example, researching a report or
designing an automotive part; management processes, which manage groups of instances of execution processes in order to report on process status, or
allocate resources to individual process instances and schedule them efficiently – for example, research scheduling or
automotive production line optimisation; and strategy processes, which work above the level of management processes – for example to set the overall policies and
priorities for management processes which will in turn manage execution process instances; or indeed to create entirely new
execution or management processes, or tune existing processes.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
BPM: beyond technology: Neil Ward-Dutton, MWD (February 2008)
|
|
|
|
|
|
Hammer & Champy’s seminal book Re-engineering the Corporation may now be an old and dusty volume, but business
processes are once again a fashionable thing to talk about. Business process management (BPM) is hot, hot, hot – every IT
vendor seems to have some kind of initiative or offering that is related to it.
But thinking about BPM purely in terms of technology is a sure way to confusion and, ultimately, failure. It’s better to think about
BPM as an approach to thinking about a business, and relating the resources that support the business to that ‘thinking tool’.
Yes, technology does play a role: both in helping to automate BPM activity, and in providing some of the resources which
support business processes. But the role of technology in BPM is complicated and you must understand it before embarking
on a BPM initiative. These articles aim to help.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
SOA: the good and bad news: Rob Hailstone, Butler Group (December 2007)
|
|
|
|
|
|
The good news is that an IT architecture built around loosely coupled services is so adaptable
that it can be used to deliver a wide variety of business benefits. The bad news (or at least the
bit that’s missing from most ‘how to do SOA’ articles) is that the same approach to adoption
doesn’t necessarily give the best results in each case.
SOA is mostly described from the strategic perspective of establishing an IT infrastructure that
is adaptable and responsive to changing requirements. This is a long-term strategic
commitment that demands a sound methodology if it is to be successful.
However, most of the deployments we see do not fit this pattern – at least in their initial
implementations. SOA is far more likely to find its way into an organisation for a much more
pragmatic reason that only makes sense if it can be delivered quickly.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
SOA: saviour or chimera?: Clive Longbottom, Quocirca (October 2007)
|
|
|
|
|
|
The mantra of the last few years has been that service oriented architecture (SOA) is another
silver bullet for solving all known IT problems, enabling IT to fulfil its dream of being the great
White Knight galloping over the hill to save the day every time there is a problem.
But just how is SOA faring in the big bad world out there? And is it just another passing fad? I
for one certainly hope not, having firmly nailed my colours to the SOA mast.
In the past I have said that SOA is the only sensible way forward for organisations wishing to
remain competitive in the medium to long term. But even I have to admit that SOA is struggling
to deliver on its promise – and I’d like to examine why this is so.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Web 2.0: now and into the future: Martin Berman, Hitachi Consulting (June 2007)
|
|
|
|
|
|
Over the last three years, a series of changes have taken place in the internet which are now
being referred to as ‘Web 2.0’. The first evolution of the web a decade and a half ago caught
many organisations by surprise. Even giants like Microsoft initially dismissed the internet. Is
history about to repeat itself? Are many organisations going to miss the opportunities presented
by Web 2.0? Are you?
This is the second of two articles. The first encapsulated what Web 2.0 is about. This article
explores the multi-faceted ways in which organisations can gain substantial competitive
advantage using this latest version of the web.
In the midst of all the hype about Web 2.0, many organisations are asking ‘So what – what can
it do for me?’ We will attempt to provide some specific answers, and examples, to that
question.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Web 2.0: now and the future: Martin Berman, Hitachi Consulting (May 2007)
|
|
|
|
|
|
Over the last three years, a series of changes have taken place in the web which are now being
referred to as ‘Web 2.0’. The first evolution of the web a decade or so ago caught many
organisations by surprise. Even giants like Microsoft initially dismissed the internet. Is history
about to repeat itself? Are many organisations going to miss the opportunities presented by
Web 2.0? Are you?
This is a series of two articles. This first article will try and encapsulate what Web 2.0 is about.
Its basic thesis is that companies have to really understand Web 2.0 and all its implications
before they can begin to use it to its full potential. Web 2.0 gives companies a whole new set of
tools. However, any tool can be dangerous if misunderstood and mis-used.
The second article will explore the multi-faceted ways in which organisations may be able to
gain substantial competitive advantage using this latest version of the web.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
SOA second sight: Steve Craggs, Saint Consulting (February 2007)
|
|
|
|
|
|
Service oriented architecture (SOA) has become a major focus in the IT industry. The market is
developing quickly – and users and potential users need as much help as possible in
understanding its future direction.
So looking at the SOA and integration marketplace as a whole, what is likely to happen during
the rest of 2007? I make six key forecasts: SOA taxes will start to bite; the SOA focus will move to application development, testing and project management; back-door SOA adoption will become prevalent; SOA business cards will emerge; the market for SOA appliances will grow; and the ESB market will consolidate.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Gaining business buy-in to SOA: Chris Maxwell, JMG Partners (April 2007)
|
|
|
|
|
|
There is a big problem with SOA. It sounds like a technical topic for technical people. It does not sound like something
that the business world could or should spend time on.
Service orientation is a concept that comes out of modern ways of thinking about IT. It is of course based on the idea
that if you can properly define the essential capabilities (or services) that you need to run your business, then these
services can be used and re-used many times, redundant services can be eliminated, disparate systems and
technologies can work together – and as a consequence you will have a far more efficient and effective operation than
you did before.
But whilst service orientation is a concept that comes from the world of IT, in practice it has even greater applicability to
business processes and business services. Indeed business service orientation should come before any consideration of
technology or systems.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Integration: still so important: Cliff Mills, PMP Research (January 2007)
|
|
|
|
|
|
Like death and taxes, the need to integrate applications and data across multiple platforms will never go away – only the most
effective way of achieving it will change.
Whatever long-term software development strategy is adopted, there will always be short-term expediency, changes in
direction and mergers and acquisitions. These can have a profound effect on an organisation’s IT and at some stage the
question “How do we link these systems together more effectively?” will need to be addressed.
A study by PMP Research on the use of integration software in over 100 major organisations shows that 85% of respondents
still see the consolidation and integration of their information systems as a highly important business challenge. On top of this,
84% believe integration software provides real benefits to their organisation – and in general the benefits are seen as far
outweighing the associated costs.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Plugging into transformation: David Ballard, Northdoor (December 2006)
|
|
|
|
|
|
According to industry analysts Evans Data Corp, .NET and J2EE (Java) are among the most
popular development platforms on the planet. A study last autumn showed that 47.9% of
developers used Java, while 51.4% developed for .NET. Some developers are clearly so taken
with the platforms that they use both. Indeed, many companies have decided that all ‘strategic’
application developments will be conducted using one of these two modern languages.
That said, plenty of organisations still have valued applications that are written in legacy fourthgeneration
languages (4GLs) such as PowerBuilder, Centura/Gupta, Oracle Forms, Visual
Basic 6, RPG and Delphi. However, skills shortages, increasing support costs, inadequate
performance and incompatibilities with system-level upgrades are exposing these companies to
risks, and increasing the cost of ownership of the legacy software.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
SOA basics: benefits to the business: Rob Hailstone, Butler Group (October 2006)
|
|
|
|
|
|
When an IT topic has been centre-stage for a couple of years, it sometimes happens that the
original purpose gets forgotten in the heat of the technical arguments. Service oriented
architecture (SOA) is at risk of falling foul of this tendency, so let’s revisit the purpose briefly
before discussing how it can be exploited.
The purpose of SOA is to enable IT to respond more quickly to changing business
requirements. This simple statement hides four decades of frustration that IT has become a
black box to the business world, with no obvious way of reconciling the time and cost of change
requests with the business value that is being sought.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Architectural power: Martin Berman, Impact Plus (August 2006)
|
|
|
|
|
|
An enterprise architecture (EA) is a blueprint or map of your entire organisation. It is ‘holistic’ in that it encompasses all aspects
of your company, such as processes, IT, locations, information and data, organisational structure and business aims and
objectives. In particular it links the business and the technology drivers – and the value of an EA arises, at least
in part, from an understanding of the linkages between all those aspects.
Impact Plus works with large, complex organisations such as the Home Office, Department for Transport, Department for
Education and Skills amongst others, and we have found that such organisations benefit significantly from developing and
maintaining an enterprise architecture.
So what benefits are to be gained? Firstly, it is important to realise that a enterprise architecture in itself has limited value. It is
the comprehensive use of the architecture that generates substantial tactical and strategic business value. And the secret to
unlocking this value is to be found in the right benefits realisation strategy, coupled with a twin-stream (tactical and strategic)
approach.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Unblocking the EAI pipes: Cliff Leach, The Project Factory (June 2006)
|
|
|
|
|
|
Enterprise application integration (EAI) using a service oriented architecture (SOA) is a hot topic
in information technology architecture around the world. Companies from Microsoft to TIBCO
and even the UK Government have invested major sums in products to support EAI and SOA.
In a nutshell, EAI/SOA architectures can be summarised as: a strong distributed heterogeneous architecture; mostly based on message-handling software backbones; used to deliver widespread, and normally asynchronous, application integration; and driving towards a federated database schema and canonical data model.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Open space: Paul Bellchambers (April 2006)
|
|
|
|
|
|
The software world is being radically changed by the open source movement. Previously, the creation of global
communities that could develop and improve a software product was not considered a practical way to bring quality
code to market. Now, however, the impact of open source is everywhere.
The concept of open source is often misunderstood, even within the computer industry, so it is not surprising that many
software practitioners and business managers are unsure as to how open source could add value to their business and
infrastructure.
Linux is synonymous with open source, but it is only one of the many software elements that are available in this way.
And a key misconception is that open source equals free software. Open source software is not free, not in the sense of
being free of cost.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
In the loop: Giffin Lorimer, G4h Ltd (February 2006)
|
|
|
|
|
|
Closed loop performance (CLP) initiatives such as balanced scorecard are an important
part of enterprise management systems. By focusing on outputs and processes, they
allow firms to organise themselves in whatever ways they need to achieve their strategic
goals. In theory this should make firms more flexible in adapting to market cycles and to
disruptive events such as technology, input costs or regulation.
The CLP approach is to break processes into measurable chunks that can be
independently improved. Part of any business improvement initiative will be a software
development task that automates the business process in order to achieve higher
consistency, faster results and lower costs.
|
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|